{"id":"arxiv/2610.37100","url":"https://arxiv.org/abs/2610.37100","tags":["math.OC"],"score":0,"title":"A Two-Stage Optimization Framework for Real-World Large-Scale University Course Timetabling Based on Orthogonalized Hierarchical Modeling and Conflict-Driven ALNS Repair","domain":"math.OC","source":"arxiv","authors":["Han-Bin Liu","Huayi Wei","Yanlin Li"],"reviews":0,"subject":"math.OC","summary":"University course timetabling is a large-scale combinatorial optimization problem involving strongly coupled temporal and resource constraints. This paper proposes a two-stage optimization framework for real-world large-scale university course timetabling, integrating orthogonalized hierarchical modeling with conflict-driven adaptive large neighborhood search (ALNS) repair. The first stage performs global course-level optimization, while the second identifies residual conflicts through a conflict graph and localizes them into connected conflict regions. The Course--Occurrence--Session…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-27","reactions":{}}